Visualization and quantification of cytotoxicity mediated by antibodies using imaging flow cytometry.

Abstract

Conventional approaches for the detection of antibody dependent cell-mediated cytotoxicity (ADCC) activity rely on quantification of the release of traceable compounds from target cells or flow cytometry analysis of population-wide phenomena. We report a new method for the direct imaging and quantification of ADCC of cancer cells. The proposed method using imaging flow cytometry combines the statistical power of flow cytometry with the analytical advantages of cell imaging, providing a novel and more comprehensive perspective of effector/target cell interactions during ADCC events. With this method we can quantify and show in detail the morphological changes in target and effector cells, their apoptotic index, the physical interaction between effector and target cells, and a directional transfer of cytosolic contents from effector to target cells. As a model system we used the therapeutic anti-CD20 antibody rituximab to target CFSE labeled Ramos human Burkitt's lymphoma cells, to CMTPX-labeled human monocytic U-937 effector cells. We expect that similar studies using different effector and target cell populations may contribute to the pre-clinical evaluation of therapeutic antibodies and help to identify mechanisms that could be beneficial in the immunotherapy of cancer.

Full text

Page 1

Research paper
Visualization and quantification of cytotoxicity mediated by antibodies using
imaging flow cytometry
Gustavo Helgueraa,b,⁎, José A. Rodrígueza,c, Rosendo Luria-Péreza,d,
Shannon Henerye, Paul Cattertone, Carlos Bregnib, Thaddeus C. Georgee,
Otoniel Martínez-Mazaf,g,h, Manuel L. Penicheta,c,g,h
aDivision of Surgical Oncology, Department of Surgery, David Geffen School of Medicine, University of California, Los Angeles, CA, USA
bDepartment of Pharmaceutical Technology, School of Pharmacy and Biochemistry, University of Buenos Aires, Argentina
cMolecular Biology Institute, University of California, Los Angeles, CA, USA
dUnit of Investigative Research on Oncological Disease, Children's Hospital of Mexico “Federico Gómez”, Mexico City, Mexico
eAmnis Corporation, Seattle, WA, USA
fDepartment of Obstetrics and Gynecology, David Geffen School of Medicine, University of California, Los Angeles, CA, USA
gDepartment of Microbiology, Immunology, and Molecular Genetics, David Geffen School of Medicine, University of California, Los Angeles, CA, USA
hJonsson Comprehensive Cancer Center, University of California, Los Angeles, CA, USA
a r t i c l e i n f oa b s t r a c t
Article history:
Received 26 October 2009
Received in revised form 6 January 2011
Accepted 10 March 2011
Available online 21 March 2011
Conventional approaches for the detection of antibody dependent cell-mediated cytotoxicity
(ADCC) activity rely on quantification of the release of traceable compounds from target cells or
flow cytometry analysis of population-wide phenomena. We report a new method for the
direct imaging and quantification of ADCC of cancer cells. The proposed method using imaging
flow cytometry combines the statistical power of flow cytometry with the analytical
advantages of cell imaging, providing a novel and more comprehensive perspective of
effector/target cell interactions during ADCC events. With this method we can quantify and
show in detail the morphological changes in target and effector cells, their apoptotic index, the
physical interaction between effector and target cells, and a directional transfer of cytosolic
contents from effector to target cells. As a model system we used the therapeutic anti-CD20
antibody rituximab to target CFSE labeled Ramos human Burkitt's lymphoma cells, to CMTPX-
labeled human monocytic U-937 effector cells. We expect that similar studies using different
effector and target cell populations may contribute to the pre-clinical evaluation of therapeutic
antibodies and help to identify mechanisms that could be beneficial in the immunotherapy of
cancer.
© 2011 Elsevier B.V. All rights reserved.
Keywords:
Antibody
ADCC
Immunotherapy
Cancer
Imaging flow cytometry
1. Introduction
In recent years the generation of therapeutic antibodies
hasexperiencedsignificantgrowth,atrendthatisexpectedto
continue, with many new developments in antibody engi-
neering focused on the improvement of antibody effector
functions (Wang and Weiner, 2008). In particular, the
mechanism of antibody dependent cell-mediated cytotoxicity
(ADCC) has been shown to play a major role in the thera-
peutic activity of rituximab (Rituxan™, Roche), trastuzumab
(Herceptin™, Roche) and other monoclonal antibodies in
development. Studies suggest that enhanced antibody
Journal of Immunological Methods 368 (2011) 54–63
Abbreviations: ADCC, antibody-dependent cell-mediated cytotoxicity;
ADCP, antibody-dependent cell mediated phagocytosis; E:T, effector to
target ratio; CFSE, carboxy-fluorescein diacetate, succinimidyl ester; DRAQ5,
1,5-bis{[2-(di-methylamino) ethyl]amino}-4,8-dihydroxyanthracene-9,10-
dione; LDH, lactate dehydrogenase; EDF, extended depth of field; FBS, fetal
bovine serum; NHL, non-Hodgkin's lymphoma.
⁎ Corresponding author at: Department of Pharmaceutical, Technology,
School of Pharmacy and Biochemistry, University of Buenos Aires, 956, Junín
Ave., 6th Floor; C1113AAD, Buenos Aires, Argentina. Tel./fax: +54 11 4964
8371.
E-mail address: ghelguera@ffyb.uba.ar (G. Helguera).
0022-1759/$ – see front matter © 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.jim.2011.03.003
Contents lists available at ScienceDirect
Journal of Immunological Methods
journal homepage: www.elsevier.com/locate/jim

Page 2

mediated effector/target cell interactions and their cytotoxic
effects on the target cells correlate with better overall
response to treatment (Dalle et al., 2008). ADCC mediates
the elimination of cancer cells through a mechanism that
requires the presence of the antibody, target cells (expressing
the antigen), and effector cells (bearing Fcγ-receptors) such
as macrophages, NK cells, monocytes, or neutrophils. In the
case of rituximab, it has been reported that leukemic target
cells opsonized with antibody can be phagocytosed by
macrophage effector cells (Glennie et al., 2007; Leidi et al.,
2009). In vitro studies that aid in the prediction of clinical
efficacy and in understanding the mode of action of thera-
peutic antibodies targeting cancer cells are of growing
interest.
Current methods for ADCC determination rely on the
evaluation of the loss of cell membrane integrity by quan-
tification of the release of traceable compounds from target
cells or by evaluation of the target cell viability by flow cy-
tometry. The51Chromium (51Cr) release assay relies on the
quantification of the radioactive material liberated from
51Cr-loaded target cells. It is a sensitive method and has been
considered the gold standard for cell-mediated cytotoxicity
(Brunner et al., 1968). However, its dependence on radioac-
tivity, the variability in labeling efficacy, and substantial
spontaneous leakage of the51Cr label in certain cell types has
limited the usefulness of this method (Jakubek et al., 1983;
Wisecarver et al., 1985). One non-radioactive alternative to
the51Crassayisthequantificationofreleaseofthefluorescent
dye calcein-AM from target cells (Metelitsa et al., 2002).
Another alternative is the colorimetric lactate dehydrogenase
(LDH) assay which measures enzyme release after disruption
of the cell membrane. It has the advantage that it is col-
orimetric and non-radioactive, but in ADCC assays this
method does not differentiate target from effector cell-
derived LDH release. Flow cytometry has also been used to
measure cytotoxicity based on the uptake of DNA fluorescent
probes after the disruption of the plasma membrane of target
cells. This method has also been used for the simultaneous
determination of ADCC and phagocytosis (ADCP) in three-
color flow cytometry assays (Karagiannis et al., 2007). Al-
though this method can effectively discriminate the viability
of the subpopulations of effector cells, target cells, and in-
teracting effector–target cells, it cannot distinguish effector–
target conjugates from events in which the effector cell has
phagocytosed the target. Fluorescence microscopy is used to
discriminate these two cases, but this technique can miss rare
events and has limited statistical power due to the limited
number of events that can be analyzed.
To comprehensively evaluate ADCC activity and effector/
target cell interactions in the presence of therapeutic anti-
bodies we used the ImageStream imaging flow cytometry
technology (Amnis Corp., Seattle, WA). This technology col-
lects multiple high-resolution images (darkfield, brightfield,
and various fluorescent) per cell in flow at high rates of image
capture, enabling statistically robust microscopy applications.
This technology has been used previously to measure simul-
taneously NK cytotoxicity and the phenotype of effector cells
(Kim et al., 2007), to determine chemically induced apoptosis
in cancer cell lines (George et al., 2004) and to study the
membrane exchange between effector and target cells in the
process of trogocytosis (Megjugorac et al., 2007; Beum et al.,
2008). Here we used this technology to monitor simulta-
neously the viability of target cells, effector cells, and to
analyze a subset of interacting effector/target cells, providing
a novel and more comprehensive perspective of ADCC events
mediated by a therapeutic antibody.
2. Materials and methods
2.1. Cell lines, fluorescent dyes, and antibody
The human monocytic cell line U-937 and the human
Burkitt's B-cell lymphoma cell line Ramos were purchased
from American TypeCulture Collection (ATCC, Manassas, VA).
Both cell lines were grown in RPMI 1640 medium (Life
Technologies, Carlsbad, CA) supplemented with 100 U/ml
penicillin, 10 μg/ml streptomycin, and 10% (v/v) heat inacti-
vated fetal bovine serum (FBS) (Atlanta Biologicals, Atlanta,
GA) at 37 °C in 5% CO2. CellTracker™ Red (CMTPX, excitation
577 nm/emission of 602 nm) fluorescent dye (Molecular
Probes, Life Technologies) was used to label the effector
cells U-937 by incubating 30 min in the presence of RPMI
1640 with 10% FBS with 2 μM CMTPX at 37 °C in 5% CO2. The
cells were washed twice and further incubated for 30 min in
RPMI 1640 with 10% FBS to remove the excess dye.
Carboxyfluorescein succinimidyl ester(CFSE green, excitation
492 nm/emission of 517 nm) fluorescent dye (Molecular
Probes, Life Technologies) was used to stain the target
Ramos lymphoma cells. Briefly, cells were washed, resus-
pended 0.1% BSA in PBS with 0.2 μM CFSE and incubated
5 min at 37 °C. The reaction was quenched by adding 5
volumes of ice-cold 5% FBS in PBS and incubated for 5 min on
ice. Cells were then washed twice with 5% FBS in PBS and
resuspended in the growth media for the experiment. The
mouse/human chimeric anti-CD20 IgG1 rituximab was
purchased from Genentech (Genentech, San Francisco, CA).
2.2. Incubation of effector and target cells with the antibody
A total of 106U-937 human monocyte effector cells labeled
with the CMTPX red fluorescent dye were incubated with
2×105Ramos human Burkitt's B-cell NHL target cells stained
with the CFSE green fluorescent dye (5:1—E:T ratio) in the
presence or absence of 5 μg/ml of rituximab for 1 h or 2 h at
37 °C in 5% CO2. We used 5 μg/ml concentration of rituximab
becauseithasbeenreportedtobeintherangetoachieveADCC
activity (Manches et al., 2003). After the incubation, the cells
were washed and fixed with 2% paraformaldehyde in PBS and
the nuclei were stained with the DRAQ5 dye (1,5-bis{[2-(di-
methylamino) ethyl]amino}-4, 8-dihydroxyanthracene-9,10-
dione, Biostatus, Ltd., Leicestershire, UK) diluted 1/200
(12.5 μM, final concentration) for 30 min. We used DRAQ5
nucleardyebecauseitcanstainnucleiofliveordeadfixedcells.
2.3. Data acquisition with ImageStream
Samples were run in the ImageStream multispectral
imaging flow cytometer (Amnis Corporation, Seattle, WA)
and images were acquired for 10,000 events/sample. Cells
were excited using a 488 nm laser with intensity ranging
from 75 to 200 mW, depending on the staining. Brightfield,
side scatter, fluorescent cell images were acquired at 40×
55
G. Helguera et al. / Journal of Immunological Methods 368 (2011) 54–63

Page 3

magnification. Only events with brightfield areas greater
than 50 μm2(excludes debris) and non-saturating pixels
were collected as previously described (Henery et al., 2008).
Imagery of excluded events was observed to prevent the
potential loss of events of interest. Single color controls were
acquired to generate a compensation matrix that was applied
to all experimental files prior to analysis.
2.4. Determination of apoptosis in effector and target cells
Apoptotic cells were identified by quantitative measure-
ment of nuclear morphology, including nuclear texture, con-
densation, and fragmentation using the IDEAS® (Image Data
Exploration and Analysis Software) package as previously de-
scribed (Henery et al., 2008). For CFSE+(target) or CMTPX+
(effector) cells, a nuclear mask was generated that contains
only pixels with intensity values in the upper 40% of the
intensity range of the DRAQ5 image. The area of this mask
(Area T40% DRAQ5) in square microns is plotted as the
abscissa in a bivariate plot against “Bright Detail Intensity R3”,
which measures the total intensity of small regions of local-
ized staining (Beum et al., 2006). Apoptotic cells with con-
densed and fragmented nuclei have low Area T40% and high
bright detail intensity values and thus can be gated on this
plot (Henery et al., 2008). This analysis was used to quantify
apoptotic CMTPX+and CFSE+events in the presence or ab-
sence of therapeutic antibody. The percentage of rituximab-
induced apoptosis in target cells in the presence of effector
cells was determined as the percentage of remaining viable
cells with rituximab as compared to cells without the anti-
body incubated in the presence of the effector cells. The
determination of percentage of apoptotic effector cells was
calculated as the percentage of apoptotic CMTPX+events
over the total number of events CMTPX+in the same
condition.
2.5. Analysis of interactions between effector cells (CMTPX+)
and target cells (CFSE+)
Images of double-positive events CMTPX+/CFSE+in all
experimental conditions were analyzed to visualize and
quantify effector–target cell contact and phagocytosis using
the IDEAS® package. To determine phagocytosis of target
cells by effector cells, we measured the distance between the
center of the CMPTX and CFSE images of double positive
events using the Delta Centroid XY (DC) feature. Events in
which effectors have phagocytosed target cells have signifi-
cantly lower DC values compared to conjugate events. In
cases where the DC is shorter than the radius of both images
reaching a 3 μm threshold, it was considered to be in the
phagocytosis/internalization event range. A histogram was
generated for the DC values of the CMTPX+/CFSE+events.
2.6. Statistical analysis
Significant differences in the data were determined by
the student's t-test using Microsoft Excel 2004 (Microsoft
Co., Redmond, WA, USA), with p≤0.05 considered to be
significant.
3. Results
3.1. Controls and assay set up
To distinguish the events corresponding to the popula-
tions of effector, target and effector–target complexes we
performed our analysis using bivariate plots. The analysis of
the treatment with rituximab compared to buffer control was
performed comparing the intensity of fluorescence in the
green channel versus the intensity of the red channel. Fig. 1
shows a dot plot of cells incubated in the presence of
rituximab where on the upper left corner U-937 CMTPX+
cells are segregated in red with an inset of the percentage of
the events in this quadrant. The arrow pointing down shows
sample imagery of events in this quadrant in red fluorescence
(CMTPX), brightfield, and DRAQ5 nuclear stain. On the
bottom right corner of the plot are gated CFSE+events in
green and a sample of events in this quadrant in green
fluorescence (CFSE), brightfield, and DRAQ5 is shown below.
On the upper right corner are gated double positive events
and its percentage over the total event count. The arrow
pointing up shows images of double positive events of
effector and target cells in contact. In dark blue are shown
CFSE+events above the threshold of CMTPX, but that are not
"true" double positive events. The Ramos cells are shown in
green, and U937 cells in red, DRAQ5 is the nuclear stain for
both cell types, and is shown also in the overlay of the
fluorescence and the overlay of brightfield, CFSE and CMTPX.
3.2. Determination of apoptosis in target and effector cells
Quantification of the apoptotic index within the popula-
tions of target and effector cells was performed by analysis of
nuclear morphology parameters such as nuclear condensa-
tion, texture, and fragmentation using the IDEAS® package.
Fig. 2A to F shows the steps that we followed to generate the
series of masks to identify an apoptotic target cell in images of
a dimeric effector/target event (Fig. 2A, B, and C). Once the
CFSE+green target cell is identified and the mask applied, we
identify the target cell nucleus stained with DRAQ5 (Fig. 2D
and E). In the region containing the nucleus of interest, a
threshold mask is then applied to the DRAQ5 nuclear image
only under the CFSE+region to determine the apoptotic
index in the target cell (Fig. 2F). Note the fragmentation and
condensation of fluorescence signal in the nucleus area with
high DRAQ5 intensity, considered evidence of apoptosis. In
contrast, we show in Fig. 2G–H images of an effector/target
event in which the target cell exhibits a homogeneous DRAQ5
staining (Fig. 2H). In this case there is no condensation or
fragmentation of the nucleus and no granularity in the
cytoplasm as would be expected inside a healthy, non-
apoptotic cell. Fig. 3 shows the scatter plot with the
morphologic metrics to determine the apoptotic index of
CFSE+Ramos target cells incubated for 2 h with effector cells
and rituximab. The gating of apoptotic and non apoptotic
events of the CFSE+population were determined based on
images of the DRAQ5 stained nucleus, where the area of the
40% threshold DRAQ5 nuclear mask within the Ramos cell
was plotted as a parameter on the Y-axis versus bright detail
intensity R3 on the X-axis. This is a “texture” feature in the
IDEAS® software package that measures the amount of
56
G. Helguera et al. / Journal of Immunological Methods 368 (2011) 54–63

Page 4

Fig. 1. Identification of effector and target cells by imaging flow cytometry. Standard dot plot analysis was used to determine red and green fluorescence intensity
in U-937 human monocyte effector cells labeled red with the CMTPX fluorochrome co-incubated for 2 h with Ramos human Burkitt's B-cell NHL target cells
labeled green with the CFSE fluorochrome (5:1—E:T ratio) in the presence of 5 μg/ml rituximab. After incubation, cells were fixed and nuclei stained with the
DRAQ5 dye. Samples were run in the ImageStream and imagery acquired for 10,000 events. In the middle of the figure we show the dot plot of the treatment with
the percentage of events CMTPX+in the upper left quadrant, CFSE+in the bottom right quadrant, and CMTPX+/CFSE+in the top right quadrant. Above the dot
plot we show representative imagery of double positive events from the upper right quadrant of the dot plot with CFSE fluorescence (CFSE green), brightfield
(BF), CMTPX fluorescence (CMTPX red), DRAQ5 fluorescence (DRAQ5), CFSE, CMTPX, and DRAQ5 fluorescence (composite), and brightfield, CFSE and CMTPX
fluorescence (BF red green). Below left we show representative imagery of single U-937 cells CMTPX positive from the upper left quadrant of the dot plot with
CMTPX red, BF, and DRAQ5 pictures. And finally, below right we show representative imagery of single Ramos cells CFSE positive from the bottom right quadrant
of the dot plot with CFSE green, BF, and DRAQ5 pictures.
57
G. Helguera et al. / Journal of Immunological Methods 368 (2011) 54–63

Page 5

remaining bright signal with a radius of 3 pixels or less after
subtraction of the detail eroded image, a parameter that is
representative of level of granularity in the mask for the
DRAQ5 image and correlates with the level of apoptosis of the
cell (Henery et al., 2008). On top of the plot representative
imagery of non-apoptotic CFSE+target cells is shown and
below the plot apoptotic CFSE+target Ramos cells are shown.
Similar procedures applied to single and dimeric CMTPX+
events allow the identification of the apoptotic effector cells,
and the discrimination between apoptotic effector and target
cells in single and double positive events to estimate more
accurately their apoptotic index.
3.3. Determination of total count and apoptotic index
The action of the effector cells in the presence of a
therapeutic antibody can result in the total destruction and
Fig. 2. Image analysis to determine apoptotic cells in double positive events.
Panels A to F show the image analysis of nuclear morphology with the
IDEAS® software used to determine the apoptotic status of a Ramos target
cell in contact with a U-937 cell. Panel A shows a double positive event in
which an apoptotic Ramos target cell is incontact with the U-937 effector cell
in an overlay of the brightfield image together with CFSE and DRAQ5
fluorescence. Panel B shows the CFSE green fluorescence of the target cell,
and panel C shows the mask created to identify the target cell. Panel D shows
DRAQ5 fluorescent staining of the event and panel E the mask containing the
fluorescent area of the condensed nucleus of the apoptotic cell. Panel F shows
the threshold that identifies only the pixel values which fall in the brightest
40% of the range of pixel values found within the nucleus of the Ramos cell.
Panels G and H show a double positive event in which the target cell is non-
apoptotic. Panel G shows the DRAQ5 staining and the mask indicating the
homogeneous nucleus of the healthy target cell, and panel H an overlay
image of the event in brightfield, CFSE and CMTPX fluorescence channels.
Fig. 3. Determination of apoptotic index by nuclear morphology. At the
center is a bivariate plot analysis using the IDEAS® software showing Area
threshold 40% DRAQ5 intensity versus bright detail intensity R3 DRAQ5
parameters of Ramos cells labeled with CFSE co-incubated for 2 h with U-937
effector cells in the presence of 5 μg/ml rituximab. The non-apoptotic Ramos
CFSE+cells are gated in green. Sample CFSE, brightfield and DRAQ5 imagery
are shown on top, with the yellow triangle pointing to the homogeneous
nucleus stained with DRAQ5 typical of non-apoptotic cells. Apoptotic CFSE+
cells are shown in orange, and representative CFSE, brightfield and DRAQ5
images are shown on the bottom, with the yellow triangle pointing to the
nucleus fragmented and intensely stained with DRAQ5, typical of apoptotic
cells. Note also in BF images the change in morphology, with intense
granularity of the cytoplasm.
58
G. Helguera et al. / Journal of Immunological Methods 368 (2011) 54–63

Page 6

fragmentation of target cells to a level that they cannot be
distinguished from debris. This activity can be quantified by
the determination of the total count of remaining healthy
target cells in the presence of effector cells and in the absence
or the presence of the therapeutic antibody as an evidence
of ADCC activity. For this purpose, a total of ten thousand
events were acquired at 2 h in the presence or absence of
rituximab at a 5:1 ratio of CMTPX+U-937 effector cells com-
pared to CFSE+Ramos target cells. The comparison was made
between antibody treated and buffer treated target cells to
discard cell killing by effector cells not mediated by antibody
and by spontaneous apoptosis of the target cells (Table 1). As
expected, in the presence of rituximab a significant percent-
age of target cells are lost compared to buffer control (Table 1,
Fig. 4A), which is interpreted as evidence of their destruction
by ADCC. Since the quantified events of the target cell popu-
lation include apoptotic cells (gray fraction in Fig. 4A), the
proportion of healthy Ramos cells is lower compared with the
total count of events and decreases as depicted in the white
area in Fig. 4A for buffer treated versus rituximab treated
cells. In the case of rituximab treated Ramos cells in the
presence of U-937 effector cells, after 2 h incubation nearly
one-fourth of the total CFSE+events detected were apoptotic
cells. In contrast, we observed that the number of apoptotic
effector cells was reduced by no more than 2% either in the
presence of rituximab or in buffer control. These observations
demonstrate that this method allows the simultaneous
quantification of the apoptotic status of effector and target
cell populations under different conditions, providing a more
comprehensive evaluation of ADCC activity. Current methods
ignore the status of the effector cells in this activity and may
overlook some potential toxicity on effector cells that might
jeopardize the therapeutic benefit of the antibody tested.
3.4. Cytoplasmic transfer from effector to target cells
Analysis of images of double positive events CMTPX+/
CFSE+also enables the observation of CMTPX+cytoplasmic
signal present within CFSE+cells (Fig. 5A). In all cases, the
overlap of the red signal was on the area of the target cells,
whichin some cases covered a large portion (N75% greenarea)
of the green signal within the Ramos cells (Fig. 5A and B). In
contrast, we did not observe cases of green cytoplasmic signal
transferred into red effectors cells, suggesting that the transfer
ofcytoplasmicmaterialoccurredfromeffectortotargetcells,as
would be expected in cases of ADCC activity. This observation
is consistent with an ADCC activity in which the antibody
facilitates this interaction between the effector and the target
cells, an activity that also occurs spontaneously, but at a lower
frequency. In the comparison between the buffer treated
Table 1
Three independent experiments showing total eventcount andpercentage of CFSE+target Ramos cells after two hour incubation withrituximab in the presence of
U-937 effector cells.
ExperimentTotal counts (CFSE+count) % CFSE+cells Cell ratio
rituximab/buffer
BufferRituximabBuffer Rituximab
1
2
3
Average
Standard deviation
t-test
7052 (1468)
10,001 (1818)
10,000 (1747)
10,000 (901)
10,000 (1625)
10,000 (961)
20.82
18.18
17.47
18.82
1.76
9.01
16.25
9.61
11.62
4.02
0.43
0.89
0.55
0.047
B
% CMTPX+ cells
0
20
40
60
80
100
120
A
0
20
40
60
80
100
120
rituximabbuffer
apoptotic
non apoptotic
% CFSE+ cells
rituximab buffer
apoptotic
non apoptotic
*
Quantification of Apoptotic Status in Effector and Target Cells
Fig. 4. Apoptotic status of target and effector cells in the presence of rituximab. Using the ImageStream imaging flow cytometer we compared the percentage of
recovery of Ramos target cells and U-937 effector cells in absence and presence of 5 μg/ml rituximab after two hour incubation time. Panel A shows the percentage
of apoptotic (gray) and non-apoptotic (white) CFSE+Ramos cells incubated with U-937 effector cells and in the presence or absence or rituximab. Panel B shows
the percentage of apoptotic (gray) and non-apoptotic (white) U-937 cells CMTPX+incubated with Ramos cells and in the presence or absence or rituximab. Error
bars indicate standard deviation of three independent experiments and (*) t-test p≤0.05.
59
G. Helguera et al. / Journal of Immunological Methods 368 (2011) 54–63

Page 7

condition and rituximab treatment, there is a significant
increase in the proportion of target cells with cytoplasmic
signalcovered 25–50% from effectorcellsinthepresenceof the
antibody (Fig. 5B). The presence of aggregates is also expected,
since the generation of homotypic aggregates of Ramos cells
mediated by rituximabhavebeenpreviously reported(Jazirehi
et al., 2007).
3.5. Image analysis of interacting effector/target cells
In order to quantify the proportion of phagocytic events
in the context of effector and target cells incubated in the
presence of a therapeutic antibody or ADCP activity, we
used the IDEAS® package to calculate the delta centroid XY.
This parameter provides the absolute distance between the
Fig. 5. Analysis of cytoplasmic transfer from effector cells to target cells in double positive events. We used an ImageStream multispectral system and acquired a
total of 10,000 events to study the double positive events containing Ramos cells stained with CFSE and U-937 cells stained with CMTPX after incubation for 1 h in
the absence and presence of rituximab. Panel A shows imagery of double positive events including CFSE fluorescence, brightfield, CMTPX fluorescence, DRAQ5
fluorescence, a composite image of fluorescent stain, and an overlay of brightfield, red and green fluorescence. In the descending rows we show representative
imagery of double positive events with increasing percentage of green (CFSE+) masked region covered by red (CMTPX+) mask, evidence of cytoplasmic transfer
from effector to target cells. At the bottom we show representative imagery of an aggregate of multiple target cells and an effector cell. Panel B shows the
percentage of CFSE+events with different percentages of CFSE area covered by CMTPX fluorescence comparing buffer treatment (open bars) and rituximab (black
bars). Each bar corresponds to different experiments and (*) t-test p≤0.05.
60
G. Helguera et al. / Journal of Immunological Methods 368 (2011) 54–63

Page 8

centers of the mask generated for the red signal to the center
of the mask of the green signal in a double positive event
(Fig. 6A). Therefore, when the distance between the centers
oftheeffector andtargetcellsislessthanhalf theradiusofthe
masks they reach a “phagocytic threshold”, in which the
effector cell is engulfing the target cell. The histogram of delta
centroid XY distribution of the double positive events allow
the estimation of phagocytic events that would be expected
between effector and target cells in ADCP activity (Fig. 6B).
We can see that of 10,000 events acquired of U-937 cells and
Ramos cells incubated for 2 h in the presence of 5 μg/ml
rituximab, only 10 events are in the phagocytic range. The
arrows pointing down show representative imagery of an
event inside the phagocytic threshold and of events beyond
the phagocytic threshold. In the case inside the phagocytic
threshold, the green signal overlaps completely with the red,
suggesting that this event may be a target cell in which
cytoplasmic material from an effector cell has been trans-
ferred, in which case we cannot count the event as true
phagocytosis. In fact, under the conditions studied we could
not detect phagocytic events in which we could observe
clearly a complete target cell inside of an effector cell. This
observation can be explained by the fact that in most of the
effector/target cell events there is some transfer of cytoplas-
mic material from the effector to the target cell and, since
these interactions are transient, we can expect to observe
some double positive events with total overlapping of
fluorophores.
4. Discussion
This method provides a more complete picture of the
events associated with ADCC activity than standard flow
cytometry. It allows the simultaneous determination of the
apoptotic index of effector and target cells, the quantification
of healthy target cells, the visualization of the characteristics
Fig. 6. Delta centroid image analysis to identify phagocytosis in double positive events. Panel A shows a composite image of brightfield, CFSE, and CMTPX
fluorescence in a double positive event of U-937 effector cells and Ramos target cells incubated for 2 h in the presence of 5 μg/ml rituximab. Superimposed is the
vector used to calculate the distance between the centers of the masks of the effector and the target cell. At the bottom of the image is the equation to calculate the
delta centroid XY. Panel B shows a frequency histogram of the delta centroid XY distances of the double positive events. We set a threshold of 3 μm for the delta
centroid XY as the phagocytic range, to identify phagocytic events. Below weshow imagery of an event inside the phagocytic range in which there is overlapping of
green and red signal. At the bottom we show imagery of events in which there is just contact of effector and target cells, resulting in the delta centroid distance
equaling the sum of the radii of the green and red fluorescent events (values well beyond the phagocytic range).
61
G. Helguera et al. / Journal of Immunological Methods 368 (2011) 54–63

Page 9

of the interaction between effector and target cells, the
formation of the cytotoxic synapse, cytoplasmic communica-
tion, and finally the quantitative analysis of rare events such
as phagocytosis of the target cell.
Treatment with rituximab, the first monoclonal antibody
approved by the Food and Drug Administration for the
treatment of human cancer, is now routinely used for the
treatment of non-Hodgkin's lymphoma. Unlike conventional
approaches for the detection of ADCC events that rely on
quantification of the release of traceable compounds from
target cells or flow cytometry analysis of population-wide
phenomena, the current work offers a novel and more
comprehensive perspective of effector–target cell interac-
tions during ADCC events. Studies in breast cancer suggest
that under physiological conditions in which high E:T ratios
are not frequent, apoptosis induction is expected to be the
main pathway of cell death mediated by ADCC (Stockmeyer
et al., 2003).
Apoptotic cell death is characterized by a series of events
that include caspase activation, cell shrinkage, membrane
blebbing, cleavage of nuclear DNA and chromatin condensa-
tion(Cohen,1993). The morphologic characteristics are easily
observed by optical microscopy, but their quantitative mea-
surement can be time consuming and difficult to reproduce
because of the lack of objectivity and throughput limitations.
We observe a reduction in the number of gated target cells in
the presence of rituximab compared to control, an evidence
of their destruction by ADCC activity. We also evaluated
the apoptotic index of the remaining cells. The detection of
Annexin V bound to phosphatidylserine exposed on the
surface of the apoptotic cells by flow cytometry is a standard
quantitative relatively unbiased method to determine the
apoptotic index. However, studies in NHL primary cells
treated with rituximab have shown that this antibody gen-
erates aggregates that exhibit high Annexin V staining, which
does not correlate with apoptosis and cannot be taken as a
reliable marker of rituximab-induced apoptosis (Manches et
al., 2003). Recent studies using automated imaging analysis
with morphology-based algorithms applied to multispectral
imagery acquired by imaging flow cytometry provide a repro-
ducible, robust and comprehensive analysis of apoptosis of a
large population of cells (Henery et al., 2008). This method
correlates highly with standard flow cytometric measurement
of apoptosis by Annexin V and caspase activation, with the
advantagethatimaginganalysisdoesnotrelyonthebindingof
a compound to a cell surface marker, allowing more accurate
exclusionofnecrotic cellsandthedetectionofearlierapoptosis
events (Henery et al., 2008). Another advantage of the analysis
usingimagesofbotheffectorandtargetcellsisthatitallowsthe
generation of digital masks to segregate clearly both popula-
tions of cells even in double positive events for accurate
quantificationand analysis of theirapoptoticindex. Incontrast,
standard flow cytometric methods such as Annexin V binding
cannot discriminate which cell is apoptotic in a dimeric
complex: a) the effector, b) the target, or c) both effector and
target cells, and assign a false positive apoptotic event to both
effector and target cell population.Wehave demonstrated that
imaging flow cytometry can be used for the determination
of ADCC activity by quantification of total number of target
and effector cells, as well as the apoptotic index of both cell
populations.
Studies of ADCC activity have shown that this event is
associated with an intimate interaction between the effector
cell and the target cell, including mutual exchange of mem-
brane lipids, the formation of the “cytotoxic synapse” (Stock-
meyer et al., 1996; Horner et al., 2007), an interaction involved
in triggering programmed cell death in the target cell (Horner
et al., 2007), and with the process of trogocytosis, in which the
targeted epitopes are shaved from the surface of the tumor cell
and transferred with membrane fragments to the effector cells
(Megjugorac et al., 2007; Beum et al., 2008). The transient
aggregate formation could be detected also in the absence of
antibody, but this interaction was extended in the presence of
anantibodytargetingtheFc receptorandthetumor-associated
antigen (Horner et al., 2007). In our case, this heterotypic
aggregation can be clearly visualized in the double positive
events between effector U-937 and Ramos cells. Although we
did not directly stain the membranes of the effector and target
cells, the imaging analysis allows us to observe transfer of
cytoplasmic content from fluorescently labeled effector cells to
the targetcells in doublepositive events,either in the presence
or absence of rituximab, with the proportion of transfer much
larger in the presence of rituximab. Other studies have shown
the generation of a cytoplasmic communication between
effector and target cells (Horner et al., 2007). Although we do
not know the nature of the cytoplasmic material transferred,
we can speculate that it may contain lytic factors that may
contribute to target cell death.
Although ADCC activity has been reported to be a major
player in the anti-tumor activity of therapeutic antibodies,
studies using M-CSF differentiated human macrophages have
shown that rituximab can mediate ADCP against human
B-chronic lymphocytic leukemia target cells in vitro (Leidi
et al., 2009). Other studies have shown that rituximab can
mediate ADCC, ADCP and apoptosis of non-Hodgkin's lympho-
ma primary cells in vitro (Manches et al., 2003). We used the
delta centroid XY analysis to estimate the frequency of
phagocytic events. Although the population of double positive
events was significant either with rituximab or buffer, the
occurrence of phagocytosis under these conditions was very
rare, and in consequence we could not quantify ADCP activity
properly. However, this study demonstrates the feasibility of
the application of the ImageStream technology to evaluate
simultaneously the ADCC and ADCP activities of therapeutic
antibodies against cancer cells in vitro. It is possible that with
othereffectorcellsand/ortargetcellsADCPeventswillbemore
frequently observed. It is expected that similar studies with
different cell populations, such as peripheral blood mononu-
clear cells, would allow the identification of different effector
cell populations interacting preferentially with the targeted
cancer cells. Another potential application of this technology
would be in a clinical setting, to identify those patients who
would be expected to be poor responders, based on poor ADCC
responses mediated by a therapeutic monoclonal antibody.
This individualized assessment would allow clinicians to tailor
treatment based on the patient's ADCC responses. It has been
reported that the clinical efficacy of cetuximab is correlated
with polymorphisms in Fcγ receptors expressed on the
patient's lymphocytes (Taylor et al., 2009), although this cor-
relation is not absolute for all antibodies and is not observed in
all malignancies (Ferris et al., 2010). This suggests that direct
measurement of ADCC may provide better prognostic
62
G. Helguera et al. / Journal of Immunological Methods 368 (2011) 54–63

Page 10

information than assessment of Fcγ polymorphisms. However,
testing this activity in primary cells from patients may be more
difficult than testing this activity against a standard cell line
such as Ramos, since the dissociation and staining of the pri-
marycancercellsmayaffecttheirsurfaceintegrityandviability,
compromising the interpretation of the results. Certainly,
further studies will be needed to confirm the clinical utility of
this technology.
In conclusion, we show that this method allows a detailed
observation of the physical interaction between target and
effector cells, their morphological changes, level of apoptosis,
and the determination of an exchange of their cytosolic
contents. Unlike conventional approaches for the detection
of ADCC events, the current work combines the statistical
power of flow cytometry with the analytical advantages of cell
imaging, providing a novel tool to better evaluate the effector/
target cell interactions during ADCC. Since this activity is
relevant also in the protection against viral infections, parasitic
and bacterial diseases (Hashimoto et al., 1983; Moore et al.,
2002), the direct visualization of the biological activity of
antibodies in the context of effector/target cell interactions
using imaging in flow cytometry, is of interest for the
characterization and pre-clinical evaluation of antibody ther-
apeutics at large, as well as for better defining mechanisms
involved in effector responses to host antibodies against target
cells.
Acknowledgments
The authors thank Dr. Tracy R. Daniels, University of
California at Los Angeles, for critically reading the manuscript,
and Raymond Kong and Ben Alderete for their technical
assistance.ThesestudiesweresupportedbytheNIH/NCIgrants
R01 CA107023, NIH/NCI R01 supplement CA107023-02S1 and
CA57152-13S1, the Howard Hughes Medical Institute (HHMI)
Gilliam Fellowship for Ph.D. studies, and the Whitcome
Fellowship of the Molecular Biology Interdepartmental Ph.D.
Program (MBIDP) at UCLA. GH is a member of the National
Council for Scientific and Technological Research (CONICET),
Argentina.
References
Beum, P.V., Lindorfer, M.A., Hall, B.E., George, T.C., Frost, K., Morrissey, P.J.,
Taylor, R.P., 2006. Quantitative analysis of protein co-localization on B
cells opsonized with rituximab and complement using the ImageStream
multispectral imaging flow cytometer. J. Immunol. Methods 317, 90.
Beum, P.V., Mack, D.A., Pawluzkowczkowycz, A.W., Lindorfer, M.A., Taylor, R.P.,
2008.Bindingofrituximab,trastuzumab,cetuximab,ormAbT101tocancer
cells promotes trogocytosis mediated by thp-1 cells and monocytes.
J. Immunol. 181, 8120.
Brunner, K.T., Mauel, J., Cerottini, J.C., Chapuis, B., 1968. Quantitative assay of
the lytic action of immune lymphoid cells on 51-Cr-labelled allogeneic
target cells in vitro; inhibition by isoantibody and by drugs. Immunology
14, 181.
Cohen, J.J., 1993. Apoptosis. Immunol. Today 14, 126.
Dalle,S.,Thieblemont,C.,Thomas,L.,Dumontet,C.,2008.Monoclonalantibodies
in clinical oncology. Anticancer Agents Med. Chem. 8, 523.
Ferris, R.L., Jaffee, E.M., Ferrone, S., 2010. Tumor antigen-targeted, monoclonal
antibody-based immunotherapy: clinical response, cellular immunity, and
immunoescape. J. Clin. Oncol. 28, 4390.
George, T.C., Basiji, D.A., Hall, B.E., Lynch, D.H., Ortyn, W.E., Perry, D.J., Seo, M.J.,
Zimmerman, C.A., Morrissey, P.J., 2004. Distinguishing modes of cell death
usingtheImageStreammultispectralimagingflowcytometer.CytometryA
59, 237.
Glennie, M.J., French, R.R., Cragg, M.S., Taylor, R.P., 2007. Mechanisms of killing
by anti-CD20 monoclonal antibodies. Mol. Immunol. 44, 3823.
Hashimoto, G., Wright, P.F., Karzon, D.T., 1983. Antibody-dependent cell-
mediated cytotoxicity against influenza virus-infected cells. J. Infect. Dis.
148, 785.
Henery, S., George, T., Hall, B., Basiji, D., Ortyn, W., Morrissey, P., 2008.
Quantitative image based apoptotic index measurement using multispec-
tral imaging flow cytometry: a comparison with standard photometric
methods. Apoptosis 13, 1054.
Horner, H., Frank,C., Dechant,C., Repp, R., Glennie,M., Herrmann,M., Bernhard
Stockmeyer, B., 2007. Intimate cell conjugate formation and exchange
of membrane lipids precede apoptosis induction in target cells during
antibody-dependent, granulocyte-mediated cytotoxicity. J. Immunol.
179, 337.
Jakubek, P., Thorsby, E., Hirschberg, H., 1983. Cell mediated lympholysis; a
modified technique using111indium-oxine-labelled targets. J. Immunol.
Methods 60, 379.
Jazirehi,A.R.,Vega,M.I.,Bonavida,B.,2007.Developmentofrituximab-resistant
lymphoma clones with altered cell signaling and cross-resistance to
chemotherapy. Cancer Res. 67, 1270.
Karagiannis, S.N., Bracher, M.G., Hunt, J., McCloskey, N., Beavil, R.L., Beavil, A.J.,
Fear, D.J., Thompson, R.G., East, N., Burke, F., Moore, R.J., Dombrowicz, D.D.,
Balkwill, F.R., Gould, H.J., 2007. IgE-antibody-dependent immunotherapy
of solid tumors: cytotoxic and phagocytic mechanisms of eradication of
ovarian cancer cells. J. Immunol. 179, 2832.
Kim, G.G., Donnenberg, V.S., Donnenberg, A.D., Gooding, W., Whiteside, T.L.,
2007. A novel multiparametric flow cytometry-based cytotoxicity assay
simultaneously immunophenotypes effector cells: comparisons to a 4 h
51Cr-release assay. J. Immunol. Methods 325, 51.
Leidi, M., Gotti, E., Bologna, L., Miranda, E., Rimoldi, M., Sica, A., Roncalli, M.,
Palumbo, G.A., Introna, M., Golay, J., 2009. M2 macrophages phagocytose
rituximab-opsonized leukemic targets more efficiently than m1 cells in
vitro. J. Immunol. 182, 4415.
Manches, O., Lui, G., Chaperot, L., Gressin, R., Molens, J.P., Jacob, M.C., Sotto, J.J.,
Leroux, D., Bensa, J.C., Plumas, J., 2003. In vitro mechanisms of action of
rituximab on primary non-Hodgkin lymphomas. Blood 101, 949.
Megjugorac, N.J., Jacobs, E.S., Izaguirre, A.G., George, T.C., Gupta, G.,
Fitzgerald-Bocarsly, P., 2007. Image-based study of interferongenic
interactions between plasmacytoid dendritic cells and HSV-infected
monocyte-derived dendritic cells. Immunol. Invest. 36, 739.
Metelitsa, L.S., Gillies, S.D., Super, M., Shimada, H., Reynolds, C.P., Seeger, R.C.,
2002. Antidisialoganglioside/granulocyte macrophage-colony-stimulating
factor fusion protein facilitates neutrophil antibody-dependent cellular
cytotoxicityanddependsonFcgammaRII(CD32)andMac-1(CD11b/CD18)
forenhancedeffectorcelladhesionandazurophilgranuleexocytosis.Blood
99 (11), 4166.
Moore, T., Ananaba, G.A., Bolier, J., Bowers, S., Belay, T., Eko, F.O., Igietseme, J.U.,
2002. Fc receptor regulation of protective immunity against Chlamydia
trachomatis. Immunology 105, 213.
Stockmeyer,B.V.T.,Repp,R.,Deo,Y.,vandeWinkel,J.G.J.,Kalden,J.R.,Gramatzki,
M., 1996. Cell adhesion molecules cooperate with Fc-{gamma}-RI in ADCC
of G-CSF primed neutrophils against breast cancer cells. Blood 88.
Stockmeyer, B., Beyer, T., Neuhuber, W., Repp, R., Kalden, J.R., Valerius, T.,
Herrmann, M., 2003. Polymorphonuclear granulocytes induce antibody-
dependent apoptosis in human breast cancer cells. J. Immunol. 171, 5124.
Taylor, R.J., Chan, S.L., Wood, A., Voskens, C.J., Wolf, J.S., Lin, W., Chapoval, A.,
Schulze, D.H., Tian, G., Strome, S.E., 2009. FcgammaRIIIa polymorphisms
and cetuximab induced cytotoxicity in squamous cell carcinoma of the
head and neck. Cancer Immunol. Immunother. 58, 997.
Wang, S.Y., Weiner, G., 2008. Complement and cellular cytotoxicity in
antibody therapy of cancer. Expert Opin. Biol. Ther. 8, 759.
Wisecarver, J., Bechtold, T., Collins, M., Davis, J., Lipscomb, H., Sonnabend, J.,
Purtilo, D.T., 1985. A method for determination of antibody-dependent
cellular cytotoxicity (ADCC) of human peripheral mononuclear cells.
J. Immunol. Methods 79, 277.
63
G. Helguera et al. / Journal of Immunological Methods 368 (2011) 54–63